Lower urinary tract symptoms (LUTS) is a highly prevalent and heterogenous syndrome that disproportionately affects older adults. LUTS can be categorized as storage, voiding, and post-voiding based on the timing of symptoms during the micturition cycle, although these categories frequently overlap within older individuals. Although pelvic floor muscle training is a well-established intervention for reducing storage LUTS by improving pelvic floor muscle strength and contraction timing, the relationship between age-related morphological changes in pelvic floor muscles and LUTS is poorly understood. It is also unknown whether these associations differ between parous women, who experience both age and birth-related changes, and nulliparous women, in whom age-related changes are not affected by pregnancy or childbirth. Prior studies primarily focused on other pelvic floor conditions, particularly pelvic organ prolapse, relied on ultrasound- based measures, and had insufficient sample size to evaluate interactions with parity status.[1] In this study, we evaluated the cross-sectional association of magnetic resonance (MR) imaging-based measures of pelvic floor morphology with LUTS, overall and by subtype, in community-dwelling older women. We hypothesized that greater urogenital and levator hiatus size would be associated with more severe LUTS, particularly stress urinary incontinence, in both parous and nulliparous older women.

The study population for this analysis included a random sample of 34 parous women and 45 nulliparous women enrolled in the Study of Muscle Mobility and Aging (SOMMA), a prospective cohort study of adults aged ≥70 years.[2] All available nulliparous women without missing data were included to maximize our ability to evaluate interactions with parity status. Eleven women were excluded due to MR artifacts that precluded pelvic morphology measurements and nine women were excluded due to missing LUTS assessment. All SOMMA participants completed a whole-body MR at baseline and 3D Slicer imaging software (https://www.slicer.org/) was used to calculate 4 reproducible pelvic morphology measures (interclass correlation coefficient >0.90 for all). Urogenital hiatus length was measured as the distance between the inferior pubic point and perineal body point. Levator hiatus length was measured from the pubic symphysis to the middle of the puborectalis bundle. Levator hiatus area was calculated in the plane of the levator hiatus as the area bounded by the pubic bone anteriorly and the lateral borders of the levator muscles as they pass behind the rectum laterally and posteriorly – this reflects the lateral bulging of the levators seen with aging. Total mid-sagittal levator area was calculated in the mid-sagittal plane as the area bounded by the pubic symphysis, perineal body, levator plate, and Sacrococcygeal Inferior Pubic Point (SCIPP) line. LUTS were assessed using the Lower Urinary Tract Dysfunction Research Network Symptom Index-10 (LURN SI-10)[3], a 10-item self-administered patient-reported outcome measure with a 7-day recall period (range 0-38; higher=more frequent). LURN SI-10 subscores include urinary incontinence (UI; urgency UI, stress UI during laughing, sneezing, or coughing, stress UI during physical activity), voiding (delayed voiding, weak urine stream), and non-UI storage (urgency, daytime frequency, nocturia). Spearman correlation coefficients stratified by parity status and partial Spearman correlation coefficients adjusted for age plus body mass index (BMI) were calculated between pelvic morphology measures and LURN SI-10 scores, subscores, or individual LUTS. Fischer’s Z-tests were used to test for interactions with parity status.

Mean age was 75±4 years for both parous and nulliparous women. Compared to parous, nulliparous women had higher BMI (27.4±5 vs 25.7±5 km/m2) and waist circumference (88±13 vs 83±13 cm) and were more likely to report lung disease (15% vs 7%). Physical activity, history of hormone replacement, multimorbidity, and polypharmacy did not significantly vary by parity status and very few women reported smoking. Parous women had an average of 2.6 pregnancies and 2.1 live births. 

Mean levator hiatus area in the plane of the levator hiatus was 63.7±8 mm in nulliparous women and 65.1±8 mm in parous women; after adjusting for age and BMI, greater levator hiatus area was moderately correlated with LURN SI-10 total score among parous women but not nulliparous women (Table 1; P-interaction=0.01). This association among parous women was predominantly driven by the voiding subscore, although weaker correlations were also observed with the storage subscore. When associations with individual UI questions were examined separately, we observed significant effect modification of the association between levator hiatus area and stress UI by parity status (Figure 1; P-interaction=0.003). Among nulliparous women, larger urogenital hiatus length and total levator area in the mid-sagittal plane were associated with lower UI subscores (Table 1). When UI types were examined separately, inverse associations with urogenital hiatus were similar for stress and urgency UI but inverse associations with total levator area were limited to stress UI during laughing, sneezing, or coughing (Figure 1). No other measures were significantly correlated with LURN SI-10 total score or subscores, although additional associations were observed with individual LUTS.

The strength and direction of associations between MR-based measures of pelvic morphology and LUTS in older women varied widely and differed by parity status. Among parous women, greater levator hiatus area in the plane of the levator hiatus was associated with higher LURN SI-10 total score, particularly voiding and non-UI storage subscores and stress UI. The direction of association was reversed among nulliparous women, greater urogenital hiatus length and total levator area in the mid-sagittal plane were both associated with lower UI subscores; these associations were primary driven by stress UI, although correlations with urogenital hiatus length were similar for urgency UI. Associations were independent of age and BMI.

Among community-dwelling older women, the relationship between urogenital or levator hiatus size and LUTS depends on parity status. Among parous older women, larger levator hiatus area is associated with increased voiding and storage LUTS and stress UI. Conversely, among nulliparous older women greater total levator area and urogenital hiatus length are associated with decreased UI. These results suggest that interventions targeting pelvic morphology to improve UI may vary parity status and, therefore, studies designed to test the effect of these interventions should be designed to account for this potential interaction.

Alshiek, J., et al., Ultrasongraphic age-related changes of the pelvic floor muscles in nulliparous women and their association with pelvic floor symptoms: A pilot study. Neurourol Urodyn, 2019. 38(5): p. 1305-1312.
Bauer, S.R., et al., Skeletal Muscle Health, Physical Performance, and Lower Urinary Tract Symptoms in Older Adults: the Study of Muscle, Mobility and Aging (SOMMA). J Gerontol A Biol Sci Med Sci, 2023.
Cella, D., et al., A New Brief Clinical Assessment of Lower Urinary Symptoms for Women and Men: LURN SI-10. J Urol, 2019: p. 101097ju0000000000000465.